key: cord-0288254-r63mmpcz authors: Frere, Justin J.; Serafini, Randal A.; Pryce, Kerri D.; Zazhytska, Marianna; Oishi, Kohei; Golynker, Ilona; Panis, Maryline; Zimering, Jeffrey; Horiuchi, Shu; Hoagland, Daisy A.; Møller, Rasmus; Ruiz, Anne; Overdevest, Jonathan B.; Kodra, Albana; Canoll, Peter D.; Goldman, James E.; Borczuk, Alain C.; Chandar, Vasuretha; Bram, Yaron; Schwartz, Robert; Lomvardas, Stavros; Zachariou, Venetia; tenOever, Benjamin R. title: SARS-CoV-2 infection results in lasting and systemic perturbations post recovery date: 2022-01-20 journal: bioRxiv DOI: 10.1101/2022.01.18.476786 sha: cc8cbc6e3ccdcf81c38b93bb3eda6998b6e9ad04 doc_id: 288254 cord_uid: r63mmpcz SARS-CoV-2 has been found capable of inducing prolonged pathologies collectively referred to as Long-COVID. To better understand this biology, we compared the short- and long-term systemic responses in the golden hamster following either SARS-CoV-2 or influenza A virus (IAV) infection. While SARS-CoV-2 exceeded IAV in its capacity to cause injury to the lung and kidney, the most significant changes were observed in the olfactory bulb (OB) and olfactory epithelium (OE) where inflammation was visible beyond one month post SARS-CoV-2 infection. Despite a lack of detectable virus, OB/OE demonstrated microglial and T cell activation, proinflammatory cytokine production, and interferon responses that correlated with behavioral changes. These findings could be corroborated through sequencing of individuals who recovered from COVID-19, as sustained inflammation in OB/OE tissue remained evident months beyond disease resolution. These data highlight a molecular mechanism for persistent COVID-19 symptomology and characterize a small animal model to develop future therapeutics. generally referred to as coronavirus-induced disease (COVID-19) (Yuki et al., 2020) . 50 In most cases among young and healthy individuals, COVID-19 is characterized by a 51 mild flu-like illness and includes limited respiratory tract congestion, fever, myalgia, 52 headache, and anosmia (Hu et al., 2021; Wiersinga et al., 2020; . 53 Amongst older populations, especially males and those with co-morbidities, COVID-54 19 can result in severe respiratory distress, multi-organ complications, and death 55 (Bhopal and Bhopal, 2020; Yuki et al., 2020) . infected lungs, peribronchiolar metaplasia could be seen expanding across large 285 areas well beyond the general vicinity of the bronchiole, thus visible even at minimal 286 magnification as a more intensely stained area (white star) ( Figure 3A ). Furthermore, 287 lungs infected by both viruses showed signs of enlarged airway spaces and residual 288 inflammation characterized by monocytes and neutrophils visible in the alveolar 289 spaces (red stars) ( Figure 3A ). This residual inflammation is in agreement with our 290 transcriptional profiling data which found Cd177 and Ly6d, neutrophil-and monocyte-291 associated genes, respectively, were significantly upregulated in lungs infected by 292 SARS-CoV-2 at 31dpi ( Figure S2A ). To confirm these findings, IHC staining was 293 these hypercellular areas oftentimes appeared co-localized with areas of lambertosis, 299 which could be distinguished via thickened alveolar walls compared to surrounding 300 healthy and mock alveolar tissues. Additionally, in line with our sequencing, which 301 identified a moderate and resolving repair response at 31dpi, Verhoeff Van Gieson 302 staining, which labels collagen and elastin fibers, showed no obvious signs of fibrotic 303 activity, collagen deposition, or elastin degradation in response to either infection 304 ( Figure S3D ). 305 Given the persistent gene signatures on day 31 post-SARS-CoV-2 infection and the 307 histological changes observed in the lung, we next assessed two distal organs via 308 H&E staining: the kidneys and heart ( Figure 3B and S3D). In the heart we observed 309 complete resolution of leukocytic infiltration at 31dpi with no noteworthy histological 310 signatures in response to infection ( Figure S3E ). In the kidney however, SARS-CoV-311 2-infected animals displayed areas of tubular atrophy characterized by thinning of 312 tubular cells and widening of the tubular lumen (black stars) ( Figure 3B ). Closer 313 examination also revealed the presence of proteinaceous fluid in the interstitial space 314 surrounding these tissues (red stars). Examination of the kidneys at this time point 315 from IAV-infected hamsters showed similar pathological findings (black stars); 316 however, the affected areas appeared smaller and less numerous than in SARS-CoV-2-infected hamsters consistent with the notion that IAV-induced damage is less 318 severe than that of SARS-CoV-2 in this small animal model. 319 To better assess the extent of infection-induced scarring, we performed quantitative 321 morphometric analyses on these histological images. Quantification of lambertosis 322 and airway size showed that these pathologies were indeed significantly greater in 323 the lungs of SARS-CoV-2-infected animals ( Figure 3C and S3F ). An identical trend 324 was also clearly visible with respect to tubular atrophy and SARS-CoV-2 ( Figure 3D ). 325 Together, these data demonstrate that both SARS-CoV-2 and IAV infections present 326 similar histological signatures in the lungs and in other peripheral organs. However, 327 despite comparable host responses, we do note a greater severity of scarring in 328 SARS-CoV-2 infection, which, given its nature, may predispose infected individuals 329 to greater functional defects in the affected organs. 330 Given that long COVID may also involve neurological and neuropsychiatric 333 symptomology (Sudre et al., 2021) , we next assessed the consequences of SARS-334 CoV-2 infection on the nervous system. For these studies, we transcriptionally 335 profiled several areas of the nervous system from 3 and 31dpi cohorts. More 336 specifically, the areas surveyed included the olfactory bulbs, medial prefrontal cortex 337 (mPFC), striatum, thalamus, cerebellum, and trigeminal ganglion (tissues collected 338 as depicted in Figure 4A ). These areas were chosen either due to their previously 339 documented positivity for SARS-CoV-2 transcripts in human patients (olfactory bulb, 340 trigeminal ganglion) or due to their functional importance in sensory, motor, cognitive, 341 and/or affective processes-all of which have been noted to be altered in subsets of hamsters infected with IAV were also collected for comparison. Following tissue 345 processing, brain regions from 3dpi were surveyed for the presence of viral RNA. As 346 expected, in hamsters infected with IAV, no viral RNA could be detected from the 347 surveyed neural tissue that aligned to the IAV genome ( Figure S4A ). In contrast, 348 within the SARS-CoV-2-infected hamster cohort, viral reads were readily detectable 349 in the nervous system in a subset of animals, consistent with the findings of others transcripts via qPCR and compared to lung, the primary site of infection ( Figure S4B -363 E). Mirroring previous data ( Figure 1A -B, S1A-B), SARS-CoV-2 sgN detection in the 364 lungs was highest on days 1 and 4. By 7dpi, sgN detection significantly diminished, 365 with only negligible levels detectable at 14dpi ( Figure S4B ). In the olfactory bulb, a 366 low level of sgN at 1dpi increased to a more prominent level at 4dpi in two of three 367 hamsters before dissipating over the next seven days ( Figure S4C Ccl11 as well as other cellular migration factor genes, such as Jaml and Rac2 545 ( Figures 5J and S6D ). T cell activation ontologies, on the other hand, were driven by 546 upregulated expression of antigen presentation markers, such as Hla-dra, Wdfy4, 547 and B2m concurrently with upregulation of T cell-associated genes such as Jak3, 548 Coro1A, Cd3e, Cd3g, and Cd3d ( Figures 5J and S6D ). In addition to immune 549 signatures, these analyses highlighted a negative enrichment for genes relating to 550 sensory perception of smell and olfaction capabilities which were present for both 551 SARS-CoV-2-and IAV-infected hamsters ( Figure S6D) . Figure S6D) . 587 To determine whether prolonged olfactory bulb inflammation was correlated with 589 altered metrics on assays that assess affective behaviors, mock-, IAV-, and SARS-590 CoV-2-treated hamsters were subjected to a marble burying assay ( Figure 6G-I) . 591 When hamsters were subjected to the this assay, an established metric for assessing 592 rodent repetitive and anxiety-like behaviors (Yanai and Endo, 2021), SARS-CoV-2-593 infected animals demonstrated a significant reduction in burying activity compared to 594 both mock and IAV groups, which performed comparably and were thus grouped 595 together ( Figure 6G ). While several studies consider increased burying as a sign of 596 chemokine, and T cell activation are presented in the graph for both IAV and SARS-921 CoV-2 compared to mock; error bars denote standard error. 922 genes with a p-adjusted value of less than 0.1 are plotted (black: p-adj > 0.05, log2 981 fold-change < 2; blue: p-adj < 0.05, log2 fold-change < 2; green: p-adj > 0.05, log2 982 fold-change > 2; red: p-adj < 0.05, log2 fold-change > 2). 983 984 (G) Lollipop charts denoting differential expression data analyzed via GSEA using the 985 Hallmark gene sets. Charts display normalized enrichment score (NES) and a dot 986 size scaled relative to -log10(FDR q-val) of the enrichment for top ten most positively 987 enriched gene sets and top three most negatively enriched gene sets in this analysis. 988 31dpi. Differential expression analysis was computed using DESeq2; differentially 995 expressed genes with a p-adjusted value of less than 0.1 are plotted (black: p-adj > 996 0.05, log2 fold-change < 2; blue: p-adj < 0.05, log2 fold-change < 2; green: p-adj > 997 0.05, log2 fold-change > 2; red: p-adj < 0.05, log2 fold-change > 2). 998 999 (D) RNA-sequencing data for all heart, lung, and kidney samples were hierarchically 1000 clustered by maximal distance between read data for each sample. Symptoms, complications and management of long COVID: a review SARS-CoV-2 genomic and 1403 subgenomic RNAs in diagnostic samples are not an indicator of active replication Small Airways Disease X-1409 linked recessive TLR7 deficiency in ~1% of men under 60 years old with life-1410 threatening COVID-19 Disrupts Splicing, Translation, and Protein Trafficking to Suppress Host Defenses QuPath: Open source software for digital pathology image analysis. Sci Rep Humoral signatures of protective 1421 and pathological SARS-CoV-2 infection in children Autoantibodies 1425 against type I IFNs in patients with life-threatening COVID-19 Pattern of cognitive deficits in severe 1429 COVID-19 Sex differential in COVID-19 mortality varies 1431 markedly by age Imbalanced Host 1434 Response to SARS-CoV-2 Drives Development of COVID-19 A standardization of the Novelty-Suppressed Feeding 1438 Test protocol in rats The effects of chronic antidepressant treatment in an animal model of 1441 anxiety The risk of new onset 1443 depression in association with influenza--A population-based observational study STAT2 1447 signaling restricts viral dissemination but drives severe pneumonia in SARS-CoV-2 1448 infected hamsters Potent neutralizing antibodies from COVID-19 patients define multiple targets of 1452 vulnerability Shotgun transcriptome, 1455 spatial omics, and isothermal profiling of SARS-CoV-2 infection reveals unique host 1456 responses, viral diversification, and drug interactions Neuropilin-1 1460 facilitates SARS-CoV-2 cell entry and infectivity The thalamus and behavior: effects of 1463 anatomically distinct strokes Striatal circuits for reward learning and decision-1466 making Neuropilin-1 1469 is a host factor for SARS-CoV-2 infection Fibroblast differentiation in wound healing 1472 and fibrosis COVID-19-related 1475 anosmia is associated with viral persistence and inflammation in human olfactory 1476 epithelium and brain infection in hamsters Brain imaging before 1480 and after COVID-19 in UK Biobank Progression in chronic kidney disease The role of medial 1484 prefrontal cortex in memory and decision making Seizures and Sleep in the Thalamus: Focal Limbic 1488 Seizures Show Divergent Activity Patterns in Different Thalamic Nuclei Peribronchiolar metaplasia: a common histologic lesion in diffuse lung disease and a 1493 rare cause of interstitial lung disease: clinicopathologic features of 15 cases COVID-19-1497 neutralizing antibodies predict disease severity and survival Importance of newly generated neurons in the adult olfactory bulb for odor 1501 discrimination Circuits between infected macrophages and T cells in SARS-CoV-2 1505 pneumonia Resting-state functional connectivity in 1508 major depression: abnormally increased contributions from subgenual cingulate 1509 cortex and thalamus Thalamus and cognitive impairment in mild 1512 traumatic brain injury: a diffusional kurtosis imaging study Pro-neurogenic effect of 1516 fluoxetine in the olfactory bulb is concomitant to improvements in social memory and 1517 depressive-like behavior of socially isolated mice Thalamic activity and 1521 biochemical changes in individuals with neuropathic pain after spinal cord injury Impaired type I 1525 interferon activity and inflammatory responses in severe COVID-19 patients Lipopolysaccharide-1528 initiated persistent rhinitis causes gliosis and synaptic loss in the olfactory bulb Neuroplastic changes in 1531 the olfactory bulb associated with nasal inflammation in mice Mouse primary microglia 1534 respond differently to LPS and poly(I:C) in vitro Olfactory 1537 bulbectomy in mice leads to increased BDNF levels and decreased serotonin 1538 turnover in depression-related brain areas Leveraging the antiviral type I 1542 interferon system as a first line of defense against SARS-CoV-2 pathogenicity Chronic 1545 anterior thalamus stimulation for intractable epilepsy SARS-CoV-2 Cell 1549 Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven 1550 Repair and regeneration of 1553 the respiratory system: complexity, plasticity, and mechanisms of lung stem cell 1554 function Immune memory from SARS-CoV-2 infection in hamsters provides variant-1557 independent protection but still allows virus transmission Cognitive impairment 1561 and altered cerebral glucose metabolism in the subacute stage of COVID-19 Characteristics of SARS-CoV-2 and 1564 COVID-19 1-year outcomes in hospital survivors with COVID-19: a longitudinal 1567 cohort study Unilateral decrease in thalamic activity observed with 1570 positron emission tomography in patients with chronic neuropathic pain Syrian hamsters as a 1574 small animal model for SARS-CoV-2 infection and countermeasure development Pharmacological 1577 evaluation of the adequacy of marble burying as an animal model of compulsion 1578 and/or anxiety The olfactory bulbectomized rat 1580 as a model of depression: an update Transcriptome assembly from long-read RNA-seq alignments with StringTie2 Enrichr: a 1587 comprehensive gene set enrichment analysis web server 2016 update SARS-CoV-2 productively infects human gut enterocytes Congenital Cytomegalovirus Infection Alters Olfaction Before 1595 Hearing Deterioration In Mice Abnormalities of thalamus volume and resting state functional connectivity in primary 1599 insomnia patients Moderated estimation of fold change 1601 and dispersion for RNA-seq data with DESeq2 Longitudinal analyses reveal 1605 immunological misfiring in severe COVID-19 A Resting-State 1609 Functional MR Imaging and Spectroscopy Study of the Dorsal Hippocampus in the 1610 Chronic Unpredictable Stress Rat Model PGC-1alpha-1614 responsive genes involved in oxidative phosphorylation are coordinately 1615 downregulated in human diabetes Post-acute 1618 COVID-19 syndrome The NF-kappaB 1621 Transcriptional Footprint Is Essential for SARS-CoV-2 Replication Neuropeptides and 1624 Neurotransmitters That Modulate Thalamo-Cortical Pathways Relevant to Migraine 1625 Headache Neurobiological 1627 signatures of anxiety and depression in resting-state functional magnetic resonance 1628 imaging MHV 1630 nucleocapsid synthesis in the presence of cycloheximide and accumulation of 1631 negative strand MHV RNA Rank-rank 1633 hypergeometric overlap: identification of statistically significant overlap between 1634 gene-expression signatures Behavioral despair in mice: a primary 1636 screening test for antidepressants Multiorgan 1639 and Renal Tropism of SARS-CoV-2 Compromised 1643 hippocampus-striatum pathway as a potential imaging biomarker of mild-traumatic 1644 brain injury and posttraumatic stress disorder Pathway enrichment analysis 1648 and visualization of omics data using g:Profiler, GSEA, Cytoscape and 1649 EnrichmentMap Cardiac involvement in the long-1651 term implications of COVID-19 Tubular atrophy in the pathogenesis of chronic kidney disease 1653 progression NIH Image to ImageJ: 25 1655 years of image analysis Thalamus structure and function determine 1658 severity of cognitive impairment in multiple sclerosis SARS-CoV-2 Nsp1 1662 binds the ribosomal mRNA channel to inhibit translation The Mesolimbic Dopamine 1665 System in Chronic Pain and Associated Affective Comorbidities Pathogenesis and transmission of SARS-CoV-2 in golden hamsters The olfactory bulbectomised rat as a model of 1672 depression Epithelial-mesenchymal transition in tissue repair and fibrosis Regulator of G protein 1678 signaling 4 [corrected] is a crucial modulator of antidepressant drug action in 1679 depression and neuropathic pain models Gene set enrichment analysis: a knowledge-based approach for interpreting 1684 genome-wide expression profiles Attributes and 1688 predictors of long COVID Bidirectional 1690 associations between COVID-19 and psychiatric disorder: retrospective cohort 1691 studies of 62 354 COVID-19 cases in the USA Isolation and 1694 sequencing of CA/GT repeat microsatellites from chromosomal libraries without 1695 subcloning Neuron-glia signaling in trigeminal ganglion: 1698 implications for migraine pathology Structural basis 1702 for translational shutdown and immune evasion by the Nsp1 protein of SARS-CoV-2 Cellular and 1705 molecular mechanisms triggered by Deep Brain Stimulation in depression: A 1706 preclinical and clinical approach Changes in 1709 thalamus connectivity in mild cognitive impairment: evidence from resting state fMRI Transmission, Diagnosis, and Treatment of Coronavirus Disease 1713 2019 (COVID-19): A Review Pathologic 1715 separation of idiopathic pulmonary fibrosis from fibrotic hypersensitivity pneumonitis Intratubular epithelial-1719 mesenchymal transition and tubular atrophy after kidney injury in mice Functional Aging in Male C57BL/6J Mice Across the 1722 Life-Span: A Systematic Behavioral Analysis of Motor, Emotional, and Memory 1723 Function to Define an Aging Phenotype A Human Pluripotent Stem Cell-based 1727 Platform to Study SARS-CoV-2 Tropism and Model Virus Infection in Human Cells 1728 and Organoids COVID-19 pathophysiology: A 1730 review Disruption of 1733 nuclear architecture as a cause of COVID-19 induced anosmia An RNA-1737 sequencing transcriptome and splicing database of glia, neurons, and vascular cells 1738 of the cerebral cortex A Novel Coronavirus from Patients with hamsters compared directly to IAV-infected hamsters at 3dpi and 31dpi via DESeq2 1057 across the sampled brain regions; differentially expressed genes with a p adjusted 1058 value of less than 0.1 are plotted (black: p-adj > 0.05, log2 fold-change < 2; blue: p-1059 adj < 0.05, log2 fold-change < 2; green: p-adj > 0.05, log2 fold-change > 2; red: p-adj 1060 < 0.05, log2 fold-change > 2). 1061 1062 (K) Rank-rank scatter plots were generated for each brain region at 3dpi and 31dpi to 1063 28dpi. Following testing with time measured to discovery of buried food, the test was 1130 also repeated with food that was visible rather than buried. Biomedicals) for 40 seconds at 6 m/s for 2 cycles in a FastPrep 24 5G bead grinder 1161and lysis system (MP Biomedicals) for plaque assay or RNA isolation, respectively. 1162Additional tissues were fixed in 4% paraformaldehyde for >72 hours prior to 1163 embedding in paraffin wax blocks for histology. Prior to fixation, lungs were inflated 1164 using 1.5 mL of 4% PFA administered via intratracheal catheter. An independent 1165 female cohort of 6-7 week-old female Golden Syrian hamsters was also obtained 1166 from Charles River Laboratories and treated in an analogous manner. These The marble burying assay was adapted from previously described protocols (Yanai 1285 and Endo, 2021). Hamsters were placed into a corner of a cage with clean bedding 1286 that had 20 equally-spaced glass marbles placed inside of it. Hamsters were allowed 1287 to move freely about the cage for 15 minutes, at which time they were moved back to 1288 their original cage. The number of buried and unburied marbles per cage were tallied 1289 by two independent observers and averaged. Partially buried marbles were counted 1290 as buried if greater than 60% of the marble was covered with bedding material. All 1291 group were assessed for outliers which were corrected for using Iterative Grubb's 1292 method. All behavioral studies were in compliance with institutional IACUC protocols 1293 and took place inside of a biosafety cabinet according to BSL-3 protocols. All autopsies are performed with consent of next of kin and permission for retention 1320 and research use of tissue. Autopsies were performed in a negative pressure room 1321 with protective equipment including N-95 masks; brain and bone were not obtained 1322 for safety reasons. All fresh tissues were procured prior to fixation and directly into 1323Trizol for downstream RNA extraction. Tissues were collected from lung, kidney, and 1324 the heart as consent permitted. Post-mortem intervals ranged from less than 24 1325 hours to 72 hours (with 2 exceptions -one at 4 and one at 7 days -but passing RNA 1326 quality metrics) with an average of 2.5 days. All deceased patient remains were 1327 refrigerated at 4˚C prior to autopsy performance.